Oxylabs is enterprise proxy and scraping infrastructure: a Web Scraper API on top of one of the largest proxy pools in the business, sold with cloud integration, batch queries, a headless browser and a custom parser included in every tier.
It is also the clearest example on this page of why a plan price is not a price.
What it costs
| Plan | Price/month | Results | Catalogue rate |
|---|---|---|---|
| Free trial | $0 | up to 2,000 | — |
| Micro | $49 | up to 98,000 | from $0.50 / 1K |
| Starter | $99 | up to 220,000 | from $0.45 / 1K |
| Advanced | $249 | up to 622,500 | from $0.40 / 1K |
| Business | $999 | up to 3,330,000 | from $0.30 / 1K |
| Custom | negotiated | — | from $0.25 / 1K |
⚠️ The "catalogue rate" column needs one clarification from us: the first four match price ÷ results, so they are our arithmetic on their figures. The Custom one ($0.25) is theirs alone — there is no result count to divide.
⚠️ The word "from" is doing a lot of work
The per-target table below is a floor, not a price, and the multiplier is the site you are scraping — this one is published on their own pricing page:
| Target | Cost per 1,000 results |
|---|---|
| Amazon | $0.50 |
| $1.00 | |
| Other sites, no JS rendering | $1.15 |
| With JavaScript rendering | $1.35 |
That is a 2.7× range inside a single plan, and social platforms land at the expensive end: they are "other sites", and they need JavaScript rendering, which is exactly the $1.35 row.
So the honest Micro-tier number for social data is not $0.50 per 1,000. It is $1.35 per 1,000 — and at that rate $49/month buys roughly 36,000 results, not 98,000.
We are not accusing them of hiding it. They published it. We are pointing out that every comparison page that quotes "$0.50/1K" is quoting the cheapest row of a table with four rows, and that the row you land on is decided by what you scrape.
Who it's for
Oxylabs is for volume and for difficulty. If you are fetching millions of pages, need residential proxies you can defend in an audit, and have an engineering team who will own the parsing, this is category-leading infrastructure and we are not a substitute.
It is not for someone who wants followers as an integer. Like
ScrapingBee, an Oxylabs result is a page — the structure
is your problem, eleven times over.
Where it is better than us
Scale, by orders of magnitude. Their Business tier moves 3.3 million results a month. We are not in that conversation.
Proxy infrastructure. They own it; we buy transit. For a target that blocks aggressively, that is the difference between a 200 and a 403.
Enterprise procurement. They have the contracts, the SLAs and the compliance paperwork a large company requires. We have a Stripe checkout.
The legal part
Oxylabs is one of the companies whose public position on scraping is substantive
rather than decorative, and the category's case law — hiQ v. LinkedIn,
Meta v. Bright Data — is the backdrop for all of us.
We publish our own line on public versus personal data and a removal route. Theirs is theirs to state, and we are not going to summarise a legal position from the outside.
Who measured this
We did, we compete in this category, and one thing about this reading is weaker than the others on this site, so it is said plainly:
oxylabs.io refused direct reading — the fetch came back
"unable to verify if domain is safe". The numbers above come from search
restricted to their own domain, including their pricing page itself. That is
a primary source, and it is one grade below reading the table on the page. We
are not going to dress it up as the same thing.
The per-source table is the part we would most want a second reading on, because it is the part that changes the answer.
What we run is an hourly archive of what's rising across public platforms. What we sell is parsed JSON in one shape — the parsing Oxylabs deliberately leaves to you.
Plan prices, result counts and the per-source cost table read from oxylabs.io/products/scraper-api/web/pricing on 2 October 2026 via search restricted to that domain, because the site refused direct fetch. Primary source, one grade below a direct page read, and flagged as such. The $1.35 JavaScript-rendering row and the 36,000-result arithmetic are the numbers to check first. Verify before deciding.